Convex High Availability?
Hey all, interested in convex since it was born really but in the last 3months its becoming more and more tempting to make the swap over.
Now I run some enterprise grade apps that have real-time requirements (currently leveraging ws) but I'm curious how I could go about self-hosting convex and if there are any guidelines for hosting it across multiple AZs or regions
My SLA req. is 99.99 (4 9's)
Or even if the Convex Cloud meets this SLA even better but I cant see anything online.
10 Replies
Thanks for posting in <#1088161997662724167>.
Reminder: If you have a Convex Pro account, use the Convex Dashboard to file support tickets.
- Provide context: What are you trying to achieve, what is the end-user interaction, what are you seeing? (full error message, command output, etc.)
- Use search.convex.dev to search Docs, Stack, and Discord all at once.
- Additionally, you can post your questions in the Convex Community's <#1228095053885476985> channel to receive a response from AI.
- Avoid tagging staff unless specifically instructed.
Thank you!
I would be surprised if you manage to self-host more reliably than the creators of Convex themselves. Planetscale, their underlying platform, is 5 9s.
And Convex themselves target 4 9s it seems: https://docs.convex.dev/production/state/
Status and Guarantees | Convex Developer Hub
Learn about Convex's production guarantees, availability targets, data durability, security features, and upcoming platform enhancements.
Understandable comment, biggest concern is its a target not a commitment in the TOS
Self-hosting is just stable from the perspective of non-breaking changes hitting my production environment as im not rolling with the SaaS GA state
The spottiness in this https://status.convex.dev/uptime/x7xtssrkpfbn?page=2 doesnt give my tech lead the confidence when im trying to convince him to shift over
Jan, Mar and July are the only times I'd be meeting contractual obligations so far this yr.
Convex Status - Uptime History
Convex's Uptime History
If i can replicate the DB or use RDS then hosting the non-db on two ec2 instances spanned across 2 AZs hit my SLA requirement (based on AWS meeting theirs obvs)
For SLA stuff recommend reaching out to sales@convex.dev
Sure ill do that then, so pretty much there isnt a "HA for self-hosting" guide available for the SH route?
I honestly don’t know on self hosting, I’d ask about that as a part of the conversation since you’re open to both
good idea, appreciated!